Total Ankle Arthroplasty Polyethylene Wear Varies with Implant Type and Mode of Failure

Introduction/Purpose: Polyethylene wear is a concern for failure of any joint replacement, including total ankle arthroplasty (TAA). Heterogeneity in bearing surface design among current TAA systems show no clear solution to the competing objectives of function (constraint and kinematics) and wear (contact stresses). Literature has begun to investigate polyethylene wear and damage; however, a comprehensive understanding of polyethylene wear patterns in vivo and location remains unknown. This study aims to quantify the type and severity of differing damage modes on the polyethylene insert from retrieved TAA prostheses following reoperation or revision. We hypothesized that polyethylene wear amount will be greater in TAAs that underwent revisions rather than reoperation, and that wear would vary between implants based on extent of constraint. Methods: This is a retrospective study of TAA patients (2007-2021) who underwent revision (removal of polyethylene and tibial and/or talar components) or reoperation (removal of polyethylene only) following primary TAA with a symmetric bicondylar (SB) implant with more constraint or an asymmetric bicondylar (AB) implant with less constraint. Demographics and surgical data were recorded. Retrieved polyethylene inserts were examined microscopically to characterize wear patterns according to a standardized protocol. Polyethylenes were divided into four regions on both the articular and backside surfaces: 1) lateral anterior, 2) lateral posterior, 3) medial anterior, and 4) medial posterior. Each region was graded by two independent raters on a scale of 0-3 based on severity for each of the following damage modes: 1) burnishing, 2) pitting, 3) scratching, 4) third body debris, 5) abrasion, 6) surface deformation, and 7) delamination. We assessed associations between polyethylene wear pattern and severity with implant type, revision, and reoperation. Results: 55 TAAs underwent revision (n=28) or reoperation (n=27). 30 (55%) ankles had primary TAA with AB implants (Salto Talaris) and 25 (45%) with SB implants (Inbone/Infinity) (Table 1). SB cohort had a shorter mean in-body duration (time from polyethylene implant to polyethylene explant) versus AB cohort (P=0.011). SB cohort had significantly greater overall polyethylene damage severity (P=0.007) and greater damage severity in all articular regions versus AB (P≤0.035 for all). Burnishing was significantly greater in SB versus AB (P < 0.001). TAAs that underwent revision had significantly greater overall damage severity versus reoperation (P=0.005), with significantly greater damage severity on articular medial posterior (P=0.003), lateral anterior (P=0.001), and lateral posterior (P=0.004) regions. Scratching (P=0.005), pitting (P < 0.001), and third body debris (P=0.036) were significantly greater in revision TAAs. Conclusion: While damage modes between SB and AB total ankle implants were similar, ankles with primary SB implants exhibited greater overall polyethylene damage severity in comparison to AB implants, despite being in-body for a shorter duration. In accordance with existing literature for total knee arthroplasty, this may suggest increased polyethylene damage severity with increased constraint. Irrespective of time in-body, failure for revision was associated with greater polyethylene damage than reoperation. This study provides the foundation for additional analyses to investigate radiographic alignment, mode of failure, and, ultimately, the association between polyethylene wear, peri-implant cysts and lucency, and TAA failure. Table 1 Patient demographics, revision, reoperation, and polyethylene in-body duration for the two patient cohorts: Patients with primary TAA with a symmetric bicondylar (SB) implant with more constraint and patients with primary TAA with an asymmetric bicondylar (AB) implant with less constraint. Values are expressed as means ± standard deviation for continuous variables, and frequencies with percentages for categorical variables. P-values reflect chi-square analyses and t tests between the two cohorts for categorical variables and continuous variables, respectively. SB cohort had significantly lower percentage of females, higher body weight, and shorter in-body duration. There was no significant difference in reoperations and revisions between the two groups.